SaaS: Too Much of a Good Thing for Retailers?
Arguably the most popular episode of "Star Trek" is Trouble With Tribbles, where a cuddly and seemingly benign creature stealthily multiplies and creates a huge problem for Kirk and his crew. Much like the Enterprise in Star Trek, retail enterprises have reached a similar tipping point with the widespread adoption of software-as-a-service (SaaS) applications.
Forrester Research claims that SaaS penetration will surpass 50 percent this year, with organizations using an average of four solutions simultaneously. Furthermore, it's expected to double in 2015. Businesses cite agility as the primary benefit of moving to SaaS, but as retailers become more widely reliant on SaaS applications, the lack of visibility and control over performance threatens the promise of agility. Can data be the difference between resilient retailing and death by a thousand paper cuts?
It Just Works
Retailers commonly run Salesforce, NetSuite, Office 365, Responsys and other SaaS apps for everything from accounting to managing direct sales agents to email marketing. Part of the beauty of SaaS is that it just works. Nowadays, you don't hear about big outages, but does that mean issues never happen?
Most SaaS vendors guarantee their service levels at 99.9 percent uptime. 99.9 percent availability translates into about 43 minutes of downtime per month. That's not a lot of time, but the more a SaaS app is used across the company and integrated, the more any interruption in service is magnified and compounded. These interruptions are like paper cuts (or tribbles), and they can add up fast. What if NetSuite slows when your accounting team is trying to close the quarter? Or when your direct sales agents are booking all their weekend trunk show orders in Salesforce.com? Or when you're launching your big holiday email campaign through Responsys?
But What Can I Do?
Any SaaS vendor worth their salt posts a "Trust" site which provides information about system status. When an outage or slowdown occurs, you'll usually find an explanation of the duration, user impact and root cause … eventually.